I assume your question 'How can you be sure a down cycle is coming.' relates to my prior comment that I expect EPS to fall below 40c/q over the next 4 Q's?
If so I base it mainly on the increased provisioning required against the loan book. This will be the main driver of lower EPS.
This is based on the following:
- Delinquencies are now pushing to the lower bound of the last recession. This means more repossessions and sales.
- Manheim index is falling. This means less money recovered on each sale.
- Dealer discount getting cut. This means less money put into the allowance for loan loss as a buffer to absorb losses. (!00% of dealer discount is put towards covering loan losses)
- Current allowance against gross financials receivables is probably getting close the limit Peter is willing to take it. This means reserve releases over the last 6Qs is coming to an end.
Based on the above I expect 1MM+ provisions to start to kick back in over the next 4Qs. This reduces EPS. Dividend should support the stock though; just means P/E expands.
One thing I noticed about the delinquency / charge off statistics is that in the early 2000's, almost 60% of the 30+ day delinquency translated into net losses over time, but that percentage has declined into a little less than 30% in recent years, and that's through the crisis. What's driving that? Have they changed how they are accounting for things or something?
I had a quick look at the delinquency reporting back in 2000 and its consistent with how they report it now. NICK are very good on the accounting disclosers and use conservative reporting. I didn't analyze the loan book back that far but my guess would be that they tightened up their underwriting when they came out of the 2000 recession -- so that may account for some of what you are seeing. Also the Manheim index went from 115 to 100 from 1999-03 causing loss rates to increase on repo sales driving higher net lossses.
I think the delinquency is what's spooking people a bit. They are having some difficulty growing loan balances, and now the delinquency rate start to tick up. But as I wrote in a previous message, it looks to me that the percent of 30+delinquency that's converting into ultimate losses (pre recovery) has been declining since early - mid 2000's, so the increase in delinquency may not be as bad as it looks at first brush. But I'm curious as to what's driving that anyway. Is that a change in how they are counting the delinquency metric? Operationally being lenient in the first month of delinquency? Or some other force at work?